Commencing Brexit

In the immediate post-referendum frenzy, public and political debate has largely revolved around the now (in)famous Article 50 of the Treaty on European Union – this hitherto unknown and untested provision is framing domestic argument about how Brexit will be commenced. The terms of Article 50 are relatively straightforward. To withdraw from the EU, the provision requires the UK to notify the European Council of its intention to do so, which triggers a period in which the Union is to ‘negotiate and conclude an agreement with that State, setting out the arrangements for its withdrawal, taking account of the framework for its future relationship with the Union’.Footnote ²⁶ The period for negotiation, conclusion and entry into force of such a withdrawal agreement is time limited at two years from the date of notification – at which point the Treaties will simply cease to apply to the departing state – although this is extendable by unanimous agreement in the European Council, and with the consent of the withdrawing state.Footnote ²⁷ There are complexities concerning what precisely Article 50 permits to be negotiated in this period: is it simply an agreement for withdrawal, or does the reference to taking account of a future framework indicate that formal negotiation of a new relationship can run in parallel?Footnote ²⁸ Whether this is practical or desirable (which is, of course, a matter of perspective) may be more significant than the broad, general terms of Article 50 – this aspect of the withdrawal process established as a matter of EU law is nevertheless likely to be the subject of further debate.

For present purposes, however, the crucial provision of Article 50 is the seemingly innocuous requirement – if indeed it can be described as a requirement, as opposed to a mere description of the inevitable – that a withdrawal decision of the kind that provides a basis for notification to the European Council is to be made by the UK ‘in accordance with its own constitutional requirements’.Footnote ²⁹ It is sensible that the government has decided to take time to establish a clearer position regarding withdrawal from the EU before triggering Article 50, which the Prime Minister has confirmed will not done be until the first quarter of 2017 at the earliest.Footnote ³⁰ This also creates time to fill with debate about what precisely the UK’s domestic constitutional requirements for a withdrawal decision might be.

It is generally agreed that the 2016 referendum itself is insufficient to constitute such a decision as a matter of law – from a legal perspective, the referendum result is ‘advisory’ rather than ‘binding’, which means that some further action is required. The government’s position appears to be that it is entitled to take the formal legal decision to leave the EU, on the basis of the ‘advice’ of the electorate as expressed in the referendum outcome, by virtue of its power to conduct foreign affairs and international diplomacy under the royal prerogative. On this view, the government could simply notify the European Council that, following the referendum result, the UK had decided to exit the EU at a moment of its choosing.Footnote ³¹

A number of arguments have, however, been developed to challenge the government’s position on triggering Article 50. First, some – most notably, Barber, Hickman and KingFootnote ³² – have suggested that an Act of Parliament is necessary to provide legal authorisation for a decision to leave the EU. The essential claim here is that a government decision under the royal prerogative would in effect render the European Communities Act 1972 – which establishes the effect of EU law in the UK’s legal system – null and void, because at the conclusion of the withdrawal negotiations under Article 50, the UK would be free of any EU treaty obligations. It would not be lawful, from this perspective, to use the royal prerogative to frustrate the purpose of legislation enacted by the sovereign UK Parliament, and therefore a statute must be enacted to authorise the government to trigger Article 50. The difficulty with this suggestion is that notification of a decision under Article 50 would not be inconsistent with, or in effect invalidate, the European Communities Act 1972. The 1972 Act is domestic legislation establishing the domestic effect of EU law within the UK – it can only be repealed or replaced by Parliament, and certainly will be during the course of the UK’s withdrawal from the EU. The notification of a decision to leave the EU under Article 50, in contrast, has no direct legal implications in the legal system of the UK – it simply instigates a period of negotiation at the European level, which follows from a process that Parliament itself legislated to initiate, when authorising the holding of a referendum on EU membership. The doctrine of parliamentary sovereignty does not require that the UK Parliament takes every decision directly; rather, it has the ultimate legislative power to establish whatever framework for decision-making it chooses.

The attempts of Arvind, Kirkham and StirtonFootnote ³³ to reach a similar conclusion – that an Act of Parliament is required to authorise notification of withdrawal – by drawing a parallel with the European Union Act 2011 also ultimately fail to convince. A decision to withdraw from the EU does not have the effect of amending or replacing the EU treaties under section 2 of the European Union Act 2011 – those treaties will be unchanged, and continue to apply to the UK until the conclusion, not the commencement, of withdrawal negotiations.Footnote ³⁴ As such, the statutory ‘referendum lock’ contained in section 2 does not come into play, and the government is not bound by the authorisation process there established. Moreover, in enacting the European Union Act 2011, the government established an extremely (probably unduly) broad scheme governing the authorisation of an array of decisions taken at EU level;Footnote ³⁵ that it explicitly chose not to bring an Article 50 decision within such a wide-ranging scheme is clear evidence that the statute should not be read to have the contrary effect, especially when we recall that the statutory purpose of the 2011 Act was to prevent power and competence being passed from the UK to the EU, rather than the reverse. The prospect of the courts taking a broader view of the requirements of the 2011 Act is extremely slim – in the leading case of Wheeler v Office of the Prime Minister the Administrative Court rejected a similar (far less politically contentious) attempt to push an expanded reading of these statutory referendum locks, and endorsed a strictly literal approach to the interpretation of the legislation.Footnote ³⁶

Once it is established that new primary legislation is not required as a matter of domestic constitutional law to authorise a withdrawal decision, a second question arises: could the power claimed by the government to exist by virtue of the royal prerogative have been displaced by a competing, existing statutory power? As Tucker has argued,Footnote ³⁷ section 2(2) of the European Communities Act 1972 may be seen to provide such a basis, in that it establishes a general power for the enactment of secondary legislation ‘for the purpose of implementing any EU obligation of the United Kingdom… or of enabling any rights enjoyed… by the United Kingdom under or by virtue of the Treaties to be exercised’. And in contrast with reliance on the prerogative power, any statutory instrument made in accordance with section 2(2) would necessarily be subject to some form of parliamentary scrutiny, and most likely, be presented by the government for the positive approval of both the House of Commons and House of Lords. Yet whether this broad statutory power is applicable to a domestic withdrawal decision, and must be used in preference to the prerogative as a result of the rule established in De Keyser’s Royal Hotel,Footnote ³⁸ largely turns on whether Article 50 (introduced only in the Treaty of Lisbon reforms in 2009) establishes a ‘right’ to withdraw from the EU, or simply creates an EU level process to acknowledge and channel a pre-existing national right held by all Member States. The vague nature of Article 50 – which for Besselink is a ‘lawyerly deformity’ which may ultimately be subject to raw politicsFootnote ³⁹ – does not make straightforward the task of working through these competing interpretations. It may be, as Elliott argues,Footnote ⁴⁰ that such an approach to section 2(2) of the European Communities Act 1972 stretches the purpose of this statutory power, which seems intended to make possible the general implementation of EU obligations in domestic law where no alternative pre-existing power is available, whereas the royal prerogative can serve to authorise a decision in the present case. The more fundamental difficulty, however, in circumstances of interpretive ambiguity, is that the government may be unlikely to accept that a right to withdraw from the EU exists ‘by virtue of the Treaties’, as would be necessary if relying on the power contained in section 2(2) of the 1972 Act. Indeed, the government would have considerable justification for regarding this as primarily a right grounded in domestic constitutional law, appropriately enough in light of the disputed language of Article 50.

The Article 50 debate in the UK is already nearing saturation point, and there is a risk that anxiety about Brexit is being projected into this single issue, distracting from, perhaps even exhausting energy for, the greater subsequent challenges. It may also be the case that attempts to shift the ultimate decision to trigger EU withdrawal to the UK Parliament – whether that takes of the form of the enactment of new legislation, or debate and scrutiny of the government’s position followed by a formal vote – is motivated, for some, by an understandable normative desire for greater parliamentary involvement at a critical constitutional moment, or even a faint hope that Brexit may yet be obstructed.Footnote ⁴¹ Yet this threatens to distort what is, in my view, a relatively clear legal position at a time when there is already more than enough constitutional uncertainty.

Crucially, however, even if the prerogative is taken to provide a legal basis for a domestic withdrawal decision, as opposed to a new Act of Parliament or an existing statutory power, Parliament will still be centrally involved in the process of UK exit from the EU. That it could be the government which commences this process, under the royal prerogative and implementing the distinct result of a national referendum, does nothing to alter the fact that Parliament will be the appropriate forum, and possesses the necessary authority, to construct the UK’s new relationship with the EU, whatever that may be, as a matter of domestic constitutional law. Though efforts will no doubt be needed to ensure that there is maximum parliamentary input into the process of reshaping the UK legal system post-Brexit – as will be discussed further below – if we approach the process holistically it becomes clear that parliamentary control over the decision to trigger Article 50 is neither legally necessary nor exclusively determinative of the extent to which Parliament will be centrally engaged in managing the change that accompanies UK exit from the EU. The legal position may be unsatisfactory, but should provide the mere starting point for the construction of the process of Brexit, which may be supplemented by the requirements of the political constitution, both principled and practical in nature.

Nonetheless, and most regrettably, we will now see the relocation of these arguments to the courts,Footnote ⁴² serving as final confirmation of the abrupt legalisation of discussion of the commencement of Brexit.Footnote ⁴³ This is in part problematic due to the inability of courts to provide real resolution to such fundamental constitutional questions as those concerning the UK’s relationship with the EU. Attempts by a succession of eurosceptics to use the domestic courts to block further integration of the UK into the EU over a number of years have rightly been criticised as opportunistic and lacking in merit.Footnote ⁴⁴ It is therefore unfortunate that it now appears that similar mechanisms are again being exploited, in what is likely to be an attention grabbing, but futile, endeavour to challenge the government’s position, unless the domestic courts seriously overstep the limits of their constitutional role. This is also problematic if the effect of a legal victory is to bolster the government and legitimise an even more executive-centred process, while perhaps functioning to crowd out non-legal arguments for some parliamentary oversight of the decision to trigger Article 50, which could have been more effective if framed in political democratic terms.Footnote ⁴⁵ The pursuit of legal challenge simply obscures this, and invites the courts into an arena where there is little chance that they can play a constructive role. Instead, the fundamental question concerning the commencement of Brexit is really ‘when?’ as opposed to ‘how?’, and this is likely to come down to an exercise of judgement by the Prime Minister and her government as to what is most advantageous for the UK’s negotiating prospects, while accommodating the European partners in whose favour the deck is already stacked.

Key challenges in execution

EU law has, of necessity, deep constitutional roots. This means the scale of the challenge presented by the idea of withdrawal from the Union is immense: it will require a combination of diplomatic negotiation at the supranational level and legislative change at the domestic level. The kind of domestic legislative change required will, of course, be contingent on the nature of the termination of membership and future relationship agreed through diplomatic negotiation, but these will not necessarily be consecutive tasks – much may need to be done simultaneously, despite the fact that the Article 50 process formally attempts to separate negotiation of an exit agreement and the full framework for a future relationship with the EU. The process of executing Brexit is therefore likely to be iterative as it develops, with diplomatic and domestic action and possibilities informed by and informing one another. While there will clearly be a great deal of difference between a future in which the UK returns to become a member state of the European Free Trade Association,Footnote ⁴⁶ or obtains some related access to the European Economic Area,Footnote ⁴⁷ and a future in which the UK stands entirely apart from the EU, and trades with the remaining member states under World Trade Organisation rules, the following section begins to anticipate some of the key challenges that will likely be experienced in any eventuality. First, we discuss who will be involved in the process of executing Brexit. Second, we will explore what kind of domestic change might be necessary.

First, who will be involved in effecting Brexit? Which UK institutional actors will have responsibility for what, and how will they interact? These questions clearly overlap with the Article 50 debate considered above, but are arguably more important, and the position is more complex. The UK government has legal and constitutional responsibility for diplomatic negotiations and foreign affairs, but EU membership also has direct domestic implications. And in the domestic sphere, the UK Parliament has the ultimate responsibility for legislating, as will be necessary to alter arrangements establishing the availability and supremacy of EU norms within the UK legal system. Of course, in many areas Parliament can only legislate to provide the framework within which the government acts, and which must be supplemented by detailed secondary legislative provision. There are important debates about whether an appropriate balance is being obtained in contemporary constitutional practice in the UK, or whether too much law-making is being left for the executive to implement, subject to lower levels of parliamentary scrutiny, via subordinate legislation.Footnote ⁴⁸ The scale of unravelling which Brexit will require will test current practice far beyond previous experience, given the range of diverse areas in which EU law will have to be replaced, from environmental, employment and consumer protection, to agricultural, fisheries and data regulation, to single market rules concerning movement of goods, services, persons, capital, and competition law. As discussed further below, such change may be carried out on an interim basis initially, but the government will inevitably lead this extensive process. There is plenty of domestic capacity which needs to be developed here: new central government departments for International Trade and Exiting the European Union have been established,Footnote ⁴⁹ and there is a need to supplement the civil service with experienced negotiators,Footnote ⁵⁰ with the operation coordinated at the highest level of government by a new Brexit Cabinet Committee chaired by the Prime Minister.Footnote ⁵¹ There are significant questions about how effectively these new teams and departments will interact, with the position of the well-established Foreign Office in Brexit less than clear, and bearing in mind the nature of some of the politicians selected by the Prime Minister to handle these sensitive, intersecting portfolios.Footnote ⁵²

There are also a range of difficult questions relating to how best the finite capacity of the UK Parliament can be exploited in this context – is the most valuable use of parliamentary time and resources to focus on deep and thorough scrutiny of the executive led review? In line with modern developments in constitutional accountability, parliamentary select committees will be key.Footnote ⁵³ Yet effort will be required to ensure coordinated activity and avoid duplication, given that many committees will have a remit which is touched by Brexit, and not just the specialist House of Lords EU Committee (and sub-committees), the House of Commons European Scrutiny Committee, and the committee which will now need to be established to scrutinise the activities of the new department for Brexit. A period of self-reflection on the role of parliamentary committees in Brexit is now well underway,Footnote ⁵⁴ but perhaps the magnitude of what is to come might be thought to justify the creation of a central Joint Committee of the Commons and Lords on UK departure from the EU, modelled on that which exists at present on human rights.

Further important questions of institutional involvement relate to the engagement of the devolved institutions with those of the UK, which has been promised by the Prime Minister.Footnote ⁵⁵ Yet Brexit arrives in an already complex environment. The national votes in Scotland and Northern Ireland in favour of remainFootnote ⁵⁶ complicate the position intensely, with discussion of a second independence referendum for Scotland in 2017 already beginning,Footnote ⁵⁷ given that the national election earlier in 2016 led to the Scottish Nationalist Party government being returned.Footnote ⁵⁸ Calls for a referendum on independence in Northern Ireland are also already occurring, though this is less easy to imagine in the immediate future,Footnote ⁵⁹ despite the fact that the land border between Northern Ireland and the Republic of Ireland could present the most difficult (and in this context, politically contentious) problem of all to resolve if a ‘hard border’ between the UK and EU was to be there established.

In Scotland, however, while there is no standing provision authorising an independence referendum, this could be very difficult to resist, as it was for the then Prime Minister David Cameron in 2012.Footnote ⁶⁰ The lack of an outright majority in the Scottish Parliament for the Scottish Nationalist Party could yet complicate things further, and there will presumably need to be a clear shift in public opinion for the First Minister Nicola Sturgeon to wish to take this on: the prospect of a second loss in an independence referendum within five years could truly serve to put independence off the political agenda for a generation.Footnote ⁶¹ Standing provision for a referendum on leaving the UK does, in contrast, exist in the Northern Ireland Act 1998,Footnote ⁶² but this power is exercisable by the UK Secretary of State, and only when ‘it appears likely to him that a majority of those voting would express a wish that Northern Ireland should cease to be part of the United Kingdom and form part of a united Ireland’.Footnote ⁶³ Well established divisions in the political class and communities of Northern Ireland would make calling such a referendum a major undertaking, even at a time when the UK did not face the all-consuming challenge of exiting from the EU. In the meantime, perhaps the crucial question is how effective engagement across the UK can be assured on an ongoing basis, even though Theresa May’s decision to deliver on the Brexit referendum vote seems to guarantee from the outset that two out of the four nations will be unhappy with the outcome, which adds to long-standing concerns about the adequacy of the existing intergovernmental machinery which is intended to smooth the operation of the devolved system.Footnote ⁶⁴ It seems probable that informal and incremental management of the internal dynamics of the UK’s devolved governance will be of great importance if anything like a ‘national consensus’ is to be achieved,Footnote ⁶⁵ and thus the manifestly political aspect of the UK’s political constitution will here be at its most evident.

Second, what kind of domestic legislative change might be necessary? The crucial domestic task necessary as a result of departure from the EU will be the repeal, and potential replacement, of the European Communities Act 1972 (along with associated primary legislation, such as the European Union Act 2011). This will remove both the direct legal effect of EU law within the UK legal system, and the domestic primacy of EU norms. Two key issues are the timing of the repeal, and whether, or with what, the European Communities Act 1972 will be replaced.

As concerns timing: the repeal will need to coincide with the formal termination of the UK’s EU membership, although there may be calls – if Brexit is to be a drawn-out process – for the domestic supremacy of EU law to be suspended in the interim.Footnote ⁶⁶ Primary legislation would be required to effect this, although given the complexity which could result, to call for such change as a matter of general principle would seem to be best understood as rhetorical posturing, rather than a worthwhile endeavour. Alternatively, where it is considered to be of particular and expedient significance, legislation on specific topics might be enacted making specific provision so as to take effect notwithstanding any contradictory EU norms – the current government may, for example, view attempts to undermine domestic guarantees flowing from the Working Time Directive as such a matter.Footnote ⁶⁷ As a matter of domestic constitutional principle, it has always been the case that in such circumstances the UK courts would be obliged to give effect to such an instruction from the sovereign UK Parliament, expressed in the appropriate statutory manner and form.Footnote ⁶⁸ With Brexit impending there can be absolutely no doubt that such a ‘notwithstanding clause’ would be legally effective. Whether the courts will respond more generally to the prevailing political atmosphere by taking a less purposive approach to the interpretation of domestic legislation to avoid inconsistency with EU law remains to be seen. It seems unlikely that we will see a clear departure from established case law like Litster v Forth Dry Dock,Footnote ⁶⁹ in which the House of Lords held that a purposive construction of domestic law would be applied (even if that involved some departure from the literal text of national legislation) where necessary to give effect to obligations under EU law. But much will depend on the stage of the Brexit process at which a case arisesFootnote ⁷⁰ – until Brexit seems an imminent certainty, and in the absence of clear legislative instructions, a formal and overt change in judicial attitudes would be surprising, although informal and implicit shifts in the courts’ approach to EU law may be inevitable.

As concerns what, if anything, will replace the European Communities Act 1972: much here depends on the nature of the UK’s future relationship with the EU. If continued access to the single market is negotiated, then domestic legislation to ensure regulatory compatibility between EU law and UK law will no doubt be required, most probably modelled on the European Communities Act 1972, if more limited in its substantive scope. This might also present an opportunity to make more explicit the mechanisms in place to remedy any inconsistency between EU and domestic law, given the almost wilful lack of clarity in the European Communities Act 1972, despite its fundamental constitutional importance.Footnote ⁷¹ In particular, the rules concerning the disapplication of incompatible UK law were developed in a haphazard judicial fashion, if inevitably so given the commitments accepted by the UK’s political institutions as part of EU membership,Footnote ⁷² and the failure (or unwillingness) of the government and Parliament clearly to articulate these consequences in statutory form.Footnote ⁷³ If such obligations are again to be accepted as part, for example, of UK membership of the European Free Trade Association,Footnote ⁷⁴ or some comparable agreement to access the European Economic Area, clearer implementing legislation must be a priority (although enduring debates about how exactly domestic ideas of legislative sovereignty can and should be preserved will surely reignite in the process).

If a looser future relationship is agreed, there may be no need to re-establish an austere imitation of the European Communities Act 1972, yet difficult legislative decisions will still be required. If the UK becomes free to legislate in a range of substantive fields previously governed by EU norms, how exactly this regained power will be exercised is a matter of real concern. It would surely be difficult to implement substantive legislative and policy change across an array of areas of domestic activity in the short term, especially when so much government time will be occupied by the formal negotiation of Brexit. It may then be that legislative frameworks modelled on EU legal norms will necessarily be preserved in the post-Brexit era, until such time that a more thorough evaluation of future possibilities can be conducted and effected. In such circumstances, provisional legislation affording domestic authority to directly effective EU rules in some areas may be required even after the UK’s membership of the EU has been terminated. This may be an unsatisfactory outcome for many on the leave side, who might wish to sever UK ties with EU law in a more definitive fashion. Yet the alternatives are also constitutionally unappealing: we either risks gaps in regulation opening up, to be discovered only at some future point as part of legal disputes or proceedings (and potentially then requiring rushed, retrospective legislative action to provide a remedy), or an executive power grab. For only the executive could have the capacity to make change across such a range of substantive areas in a short space of time, and a government could perhaps on that basis persuade the UK Parliament to make wide ranging provision for the availability of secondary legislative powers to ‘replace EU law’. Many will feel that a government intending to rewrite vast (and complex) swathes of the domestic legal system should first be required to submit such broad plans for approval by the electorate at a general election – that there is a democratic mandate for Brexit, by virtue of the referendum result, is clear, yet there is no such mandate for the government unilaterally to fill the policy vacuums potentially created by the withdrawal of EU law. That any policy choices made during this period would be reversible in future does not diminish the fact that it would be difficult for Parliament to provide adequate political oversight of so much so quickly, and in any event, interim solutions can readily acquire a degree of durability once in place.

A further complication is that withdrawal from the EU will almost certainly necessitate the amendment of the UK’s devolution legislation, and this is a very sensitive constitutional task. In particular, EU law provides a crucial statutory limit to the legislative competence of the devolved legislatures: the Scottish Parliament,Footnote ⁷⁵ the Northern Ireland Assembly,Footnote ⁷⁶ and the National Assembly for Wales.Footnote ⁷⁷ These statutory limits will presumably need to be removed or altered (depending on the precise nature of the UK’s future relationship with the EU), and while there is no legal barrier to the UK Parliament amending the terms of legislative devolution, a constitutional convention – known as the ‘Sewel convention’ – exists which provides that this will usually be done with the consent of the devolved institutions.Footnote ⁷⁸ If this is deemed to be necessary,Footnote ⁷⁹ whether devolved legislative consent can be obtained, especially in relation to the remain voting Scotland and Northern Ireland, and even given the pro-remain Labour government of Wales, will likely depend on the degree to which the UK is to separate itself from the EU. The convention has also been bolstered in relation to Scotland through its articulation as a principle which will ‘normally’ be respected in the most recent (pre-Brexit) statutory amendments to the devolution system.Footnote ⁸⁰ The framing of this statutory provision does not have the effect of making a failure to comply with the Sewel convention unlawful, but does seem to enhance the formal authority of what was already a very significant and well-respected political precept underpinning the UK’s system of devolved governance.Footnote ⁸¹ Of course, if Brexit produces an independence referendum in Scotland (or even Northern Ireland) then the UK government will have much bigger problems than recrafting the devolution Acts. Yet even in the absence of this most radical of challenges to the very existence of the UK as currently constituted, it is clear that devolution will complicate the UK’s withdrawal from the EU in ways that offer no obvious resolution capable of reconciling the competing interests at stake.

Concluding and legitimising the process

Once a deal is in place to effect the exit of the UK from the EU, the terms are known and domestic legislative changes have been agreed (at least provisionally) to come into force upon the termination of membership, how will the process be concluded? A crucial issue in this sense is whether some further legitimation of the terms of Brexit is required as a matter of domestic constitutional law or principle. The Brexit referendum delivered a narrow, yet clear and decisive, mandate for the UK to leave the EU. But the nature of that exit (to put the point in the most charitable way) could not be known in the early months of 2016, and in any event (to put the point less charitably) there was a manifest failure by the major leave campaigners to articulate any kind of clear, coherent or consistent vision of the UK’s future outside the EU. Is, then, some further resort to the electorate necessary to give a decision as momentous as departure from the EU the kind of legitimacy that the UK’s democratic political constitution ought to be understood to require?

There have already been calls for a second national referendum to confirm the outcome of the Brexit negotiations.Footnote ⁸² The UK’s uncodified constitution contains no legally binding rules which prescribe when it will be necessary for a referendum to be held; instead, arguments for a referendum must be framed and justified in political constitutionalist terms.Footnote ⁸³ Whether a persuasive argument can be so constructed is likely to remain a matter of serious debate in the coming years. From one perspective, the UK’s relationship with the EU is clearly of sufficient constitutional significance to justify direct democratic decision-making. And if the decision to leave was taken by the electorate, it is similarly arguable that the terms of departure are also a matter for ‘the people’ of the UK. Yet there are also a range of problems with this position. First, at the conclusion of negotiating Brexit, the point may well be moot – there may be no way back with our European partners after potentially punishing exit discussions, and the satisfaction or otherwise of the UK population with the terms of the final deal on offer may be utterly irrelevant.Footnote ⁸⁴ If a superior deal is unlikely to be obtainable, could the UK simply withdraw the Article 50 notification and retain EU membership? The legal possibility of this is open to question,Footnote ⁸⁵ but to seek to remain at that point would bring accusations of frustration of the original Brexit mandate, and could further inflame UK political debate, potentially ensuring even greater anti-European sentiment in perpetuity. This in itself indicates a second difficulty: a further referendum risks becoming a vehicle for the perpetuation and aggravation of a range of existing resentments focused at both domestic and European political actors, and triggering accusations (whether justified or opportunistic) of elite rejection of popular decision-making, which is of course not an unknown phenomenon when it comes to direct democracy and the EU.Footnote ⁸⁶ There is emerging evidence that the result of this referendum can be understood as part of a broader ‘anti-politics’ narrative, and that suspicion of domestic, European and international elites drove many who rejected the status quo.Footnote ⁸⁷ The disjunction between political elites and the broader electorate is a profound constitutional problem – not just in the UK or the EU, but across democratic states – and in responding to this challenge, action which has the potential to exacerbate the situation should be avoided.

Third, even if judged to be worth the potential risks, a second referendum to approve the terms of Brexit creates serious potential for clashing direct democratic mandates. Could popular rejection of the proposed future EU relationship be interpreted as a vote to remain? Even if a referendum question could be framed to avoid this possibility, that may miss the point – for many advocating a second referendum, the reversal of the decision to leave is exactly the goal of going back to the electorate. Yet this path takes us into very difficult constitutional territory; indeed, it almost invites a conflict which is irresolvable on its own terms, and it is not as if the UK is lacking for such conflicts already. Finally, would a failure to exit now on the basis of some alternative direct democratic mandate simply condemn the UK (and indeed the other member states) to experience ‘neverendums’ on continuing EU membership? Not only might this obstruct EU plans for reform and redevelopment,Footnote ⁸⁸ but also inhibit future UK governments from attempting to articulate and establish a more positive relationship with Europe. That the latter could even be attempted might seem difficult to imagine in the present environment, but the reductive defence of EU membership and its national benefits which dominated mainstream debate during the Brexit referendum suggests that fresh thinking may be required as to the UK’s aspirations regarding its relationship with Europe as a continent, as well as a political entity.

If the idea of a second referendum would pose difficulties which are, in my view, insurmountable, could there be an alternative way of legitimising the Brexit negotiations? We may consider whether a domestic general election could have some role to play in the process of exiting the EU. To require a UK government making such momentous decisions to seek a specific mandate for its actions could have great democratic value and importance. Yet the crucial question is when this could be done. Given the lack of clarity about what Brexit might amount to exhibited during the referendum campaign, and the spectrum of potential outcomes, I would argue that it is constitutionally necessary for a government to obtain a specific mandate approving the kind of relationship with the EU that it wishes to negotiate before attempting to make that vision a reality.Footnote ⁸⁹ Yet the possibility of this seems diminished at present – although far from an absolute barrier, recent changes to the rules concerning ‘early general elections’ which accompanied translation of the power to dissolve Parliament from the royal prerogative into primary legislation, make such a prospect more complicated,Footnote ⁹⁰ and the current Prime Minister appears to have no desire to test the new statutory process.Footnote ⁹¹ More crucially, however, the utility of a general election may be clear at the start of the process – in providing a test of the agenda for negotiation set by the government – but it would be limited at the end, if intended to provide an opportunity to approve a Brexit deal. Introducing a new UK government at this point, with some potentially alternative indirect (as opposed to direct) democratic mandate, if even plausible at all, would generate unparalleled uncertainty, and revisiting commitments made in good faith would surely be unpalatable to the remaining member states.Footnote ⁹² Instead, the realpolitik would surely be that such a new government would have at best only limited room for manoeuvre, and Brexit would have pushed past the point of no return.

The questions of legitimacy which revolve around Brexit are, of course, fluid and (rightly) contested. Further debate may reveal new options, or the difficulties I have traced here may be overcome by events, making either a second referendum or general election at some point a more foreseeable possibility. Yet perhaps we should also hesitate before linking questions of legitimisation exclusively to the conclusion of negotiations on exit from the EU. The very contestability of the legitimacy issues associated with Brexit means that they may also pervade other stages or elements of the process in as yet unpredictable ways. If we were being hyper-optimistic (and probably unduly so) we might consider whether the execution of Brexit could be approached in innovative ways which enhance the legitimacy (and potentially quality) of some of the many decisions to be taken concerning the replacement of substantive bodies of EU law with domestic rules. It could be possible to go beyond thinking about a review by a Royal Commission composed of the great and the goodFootnote ⁹³ – reinforced parliamentary mechanisms would be preferable in any event, in order to avoid the poor political optics associated with such elite level panels – and instead try to frame such processes around a more democratic impulse. For example, further evidence of the utility of direct citizen involvement in deliberative decision makingFootnote ⁹⁴ might suggest that there could be real value in some form of citizen-led convention – one recent version of which was employed to examine proposals for constitutional change in Ireland from 2012 to 2014Footnote ⁹⁵ – to establish a direction for UK legal reform in the areas potentially left vacant by the withdrawal of EU norms. That Brexit is of sufficient constitutional significance and scale to justify such a complex understanding – supported and structured by expert input – seems obvious, and this could be a constructive way of exploiting existing popular engagement with the idea of the UK’s post-EU future. However, it seems most unlikely that the government would be willing to risk limiting its unique opportunity to lead a reshaping of UK law after the responsibility for so doing has been prised back from the supranational level. As the idea of direct citizen involvement in executing Brexit verges on the utopian in a potentially dystopian period of uncertainty, it is hard, as a result, to see it gaining any traction regardless of its potential merits.

Exposing limitations of the UK’s constitutional arrangements?

The range and extent of the challenges that Brexit potentially creates for the UK constitution have been examined above (although inevitably incompletely, at this early stage of the process). Across the range of potential problems – which include those concerning the requirements to trigger Article 50, the respective roles of the UK Parliament and government, the position of the devolved institutions and the future of the territorial state, and the domestic legislative changes necessary – a number of patterns might appear to emerge. In different ways, we see a lack of clarity about what should happen, an array of competing actors to engage and satisfy, a multiplicity of principles to aspire to respect, and uncertainty about what the consequences of Brexit might ultimately be.

In such circumstances of pervasive difficulty, we might be led to ask whether Brexit is here serving to expose limitations in the UK’s constitutional arrangements or constitutional model. If we look beyond the throwaway normativity of claims that the UK has no constitution at all,Footnote ⁹⁷ there is a distinctive form of constitutionalism classically understood to be in operation. In the absence of a definitive constitutional text or instrument, the UK’s uncodified constitution affords significant space for political action. This vision of the political constitution is defined by a sovereign UK Parliament with the flexibility to reshape the constitution by primary legislation, a fusion of executive and legislative power underpinned by ideas of ‘responsible government’ procured through electoral and parliamentary accountability, and relatively peripheral judicial oversight operating principally where the boundaries of allocated power have been exceeded by public authorities. Against the backdrop of Brexit, it is not unreasonable to question whether the UK constitution’s very openness to politics is also its fundamental systemic flaw, hindering the ability of the various institutional actors to organise a coherent response to the challenges presented. Indeed, that it will be left for these various actors to organise the response to Brexit, due to a distinct lack of pre-existing constitutional prescription as to the domestic path by which to depart from the EU, may in itself be taken to be the ultimate failing of the UK’s political constitution.

There is clearly something to be said in support of this line of argument. It is certainly the case that Brexit has potential both to illuminate and exacerbate a range of existing ongoing difficulties concerning the UK’s contemporary constitutional arrangements.Footnote ⁹⁸ The challenges for the UK constitution which have been explored above do not arise in a vacuum; instead, to a significant extent they reflect pre-existing difficulties in a constitutional system which is already very much in flux.Footnote ⁹⁹ Concerns about the division of power between the legislature and the executive, the extent of constructive engagement between the institutions of UK central government and the devolved governments, the place of direct democratic decision-making, the effectiveness of mechanisms of parliamentary accountability, the appropriate balance between primary and delegated law-making, and the future direction of the United Kingdom itself, are not new – rather they are given fresh impetus by Brexit. It is, moreover, unlikely that additional attention or aggravation will increase the prospect of broader domestic constitutional reform. If the process of departing from the EU becomes politically and constitutionally all-consuming, it may be enough to hope that these difficulties can simply be managed, rather than expecting solutions to emerge. In that sense, Brexit may provide a case study which demonstrates acutely our existing constitutional inadequacies.

Equally, however, there is room to be sceptical of the assumptions which appear to underpin claims that Brexit has exposed the limitations of the UK’s present constitutional model. First, the UK constitution has changed, is still changing, and this is in no small part a direct result of membership of the EU. It is a matter for debate how far the political constitutionalist paradigm still accurately represents the modern position in the UK.Footnote ¹⁰⁰ We must also be careful not to present a vision of the UK constitution which ossifies certain principles, arrangements or philosophies. Nevertheless, in my view, the political constitutionalist paradigm does still have clear salience in the UK context – both descriptively and normatively – despite the changes which have occurred. Indeed, that this constitution is continually open to reformation is (at least ostensibly) one of its defining characteristics, and the extent of change in recent history can be understood as a vindication of the continuing force of the idea of a political constitution. But this does not mean that the openness and flexibility of the UK constitution should be overstated, nor that it is chronically uncertain, or even descends into ‘incoherence’.Footnote ¹⁰¹ There are actually areas where the required position is as clearly established as might be expected in a more formal constitutional order – the debate as to whether the UK government has the power to trigger notification of intention to withdraw from the EU for the purposes of Article 50 under the royal prerogative is a good example of this. As discussed above, here we see a relatively settled position being unsettled by some (though certainly not all participants in the academic debate) who may wish to cultivate constitutional arguments which can be wielded (particularly through legal action) to inhibit, obstruct or delay the Brexit process. There may be a danger in accepting the argument that Brexit has revealed the UK constitution to be fundamentally flawed, if it allows the superficial openness of the constitution to be exploited to establish an environment in which everything can be presented as remaining up for grabs. This misrepresents the true position, according to which the parameters of a domestic constitutional response to Brexit can at least be mapped and understood, even though many details await definition.

Second, we must bear in mind the unprecedented nature of Brexit. Any constitution would face challenges in dealing with such a major, unanticipated recalibration of its political architecture. And any constitution would struggle to meet those challenges, as seems likely for the UK constitution (and indeed, the ‘written’ constitution of the EU) as the process progresses. Yet it is nevertheless arguable that the political nature of the UK constitution, in addition to ensuring that the difficulties experienced will be brutally evident, will also make apparent what is surely the only way in which they can be confronted – through coordinated political action. In this sense, the UK will not be unduly constrained by law in responding to this momentous national challenge. Indeed, where we currently see legal means of problem-solving being emphasised – as regarding the domestic legal basis for giving notification of departure in accordance with Article 50 – this is likely to introduce additional, and unnecessary, complexity. For not only does this legal challenge have the potential to delay the process of Brexit, but in framing the commencement debate as one which concerns what is required as a matter of law, rather than what is desirable as a matter of political principle, the parameters of what might be conceivable have been regrettably narrowed. In contrast with what is likely to result from this legal parody, the real decisions to be made at this point are political. Moreover, it will be political pressure, incorporating both practical and principled concerns, and utilised by all manner of interested stakeholders, which is likely to be the most effective way of shaping the Brexit process.

We should not, therefore, reject the political constitutionalist model in reaction to Brexit – indeed, while existing difficulties with the UK’s specific constitutional arrangements are likely to be magnified through the process of departing from the EU, the political constitution may offer the best hope (or indeed the only hope?) for an effective response to the challenges we can expect to experience. Such a response will be one that recognises that political compromise must be at the heart of organising the process of Brexit – both at the supranational level, between the UK, the EU institutions and remaining member states, and at the domestic level, between the various institutional actors operating at the centre of government and across the UK, and the broader electorate who have set us on this path. While success cannot be guaranteed – indeed, it is hard at this point to know what ‘success’ would even look likeFootnote ¹⁰² – that the UK constitution does not conceal the essential nature of what lies ahead suggests that there is virtue to be found in the honesty of a constitutionalism which embraces, rather than simply tolerates, its fundamental politics.

However, in the political constitutionalism which is likely to dominate the UK’s response to Brexit, can we also see some of the root causes of the dissatisfaction motivating the decision to depart from the EU? Should Brexit also be understood as a challenge of the UK constitution because the result of the referendum was – in part – a function of constitutional considerations?

UK constitutionalism as a contributory factor in generating Brexit?

To explain in detail the causes of the UK’s decision to exit the EU is clearly a task which extends well beyond constitutional analysis. It is also a task which will take time, although there has already been much debate as to the initial indications. Clearly significant was the fact that the commitment to freedom of movement required by EU membership clashed with concern about immigration levels which were cynically, even viciously, exploited by some prominent leave campaigners.Footnote ¹⁰³ Electoral demographics obviously had an important impact, particularly the enhanced engagement of older votes,Footnote ¹⁰⁴ as did the failure of the then Prime Minister David Cameron to bring his party with him on a remain vote.Footnote ¹⁰⁵ Yet in addition to these diverse contributing factors we have reason to question whether there was a constitutional element to the result of the Brexit referendum.

Was the nature of UK constitutionalism itself a factor? As the campaign developed, it was notable that constitutional themes and ideas seemed to resonate in the political and public discourse to a greater extent than might have been expected, notwithstanding the fact that whether the UK should have retained its membership of the EU is obviously a ‘constitutional’ question with ‘constitutional’ implications. This does not, of course, mean that constitutional considerations dominated the debate. Yet as seemingly well-established strategies for electoral success – most notably the ‘it’s the economy, stupid’ approachFootnote ¹⁰⁶ – failed ultimately to persuade a majority of the electorate to maintain the status quo, the leave campaign’s pseudo-constitutionalist call to ‘take back control’ was effective in capturing and combining a range of issues. If we look beyond the superficiality of this message, we may discern a number of constitutional dynamics potentially at work in the Brexit vote.

The constitutional underpinnings of the referendum result – or of the long-cultivated eurosceptic environment in which the decision to exit the EU was reached – may include, among others, a discomfort with the ‘eurolegalism’ which has arguably come to define the European project,Footnote ¹⁰⁷ and the resulting changes in domestic law and practice which such a supranational philosophy prompts. A related concern may be that the UK’s domestic constitutional reliance upon mechanisms of governmental accountability and responsibility (imperfectly though they may operate in practice) does not readily translate to the supranational level of governance.Footnote ¹⁰⁸ Such discontinuities between organising principles of EU and UK constitutionalism can of course be overstated, and some degree of divergence has been inevitable by virtue of being a member state. Yet an underlying tension between these different visions of political organisation – ‘national’ and ‘European’ – may still remain, or even merely influence UK perceptions of the EU. Indeed, a general scepticism towards the power concentrated within remote international or supranational governance networks, of which the EU is the most advanced kind, may be a constitutional attitude which – whether justified in these particular circumstances or not – could have contributed to a rejection of UK participation in this architecture of European cooperation.

Perhaps the most significant constitutional idea to feature in Brexit debates, however, was the notion of sovereignty. Exactly what was meant by this seemed to vary from speaker to speaker, and from day to day.Footnote ¹⁰⁹ In some ways it may be an umbrella term capturing some or all of the potential concerns identified above: a preference for national level decision making through domestic processes, over pooled power and European influence. The vagueness of the idea of sovereignty deployed during the Brexit referendum differs markedly from the domestic idea of the sovereignty of the UK Parliament, which is a foundational, symbolic principle of UK constitutionalism.Footnote ¹¹⁰ Yet it might nonetheless be thought that the very primacy afforded to this notion of sovereignty in the domestic constitution invites, or at least facilitates, a preoccupation with questions as to the location of ultimate authority which was evidently exploited – in a rough and ready fashion – in arguments about Brexit. That there exists broad consensus that the UK conception of parliamentary sovereignty had been reconciled successfully with the core precepts of EU membership (and in particular, the domestic supremacy which must be afforded to EU law) did not seem to be significant during the course of the referendum campaign. Instead, the notion of sovereignty employed to justify Brexit was more emotive and less precise, leaving it open to debate exactly what those citizens who were persuaded by such arguments were voting to restore.

It is not my purpose to endorse these claims as to the (potential) constitutional underpinnings of Brexit, nor to condemn the domestic reliance on the idea of parliamentary sovereignty as a central organising concept of the UK constitution. Instead, I simply want to consider: if there were constitutional factors which contributed to the referendum decision to leave the EU, such as those sketched above, can they be responded to? Or are these really constitutional concerns which extend beyond the UK’s relationship with the EU? This will be important in two ways: first, regarding the future development of the UK constitution, and second, regarding the future of the post-Brexit EU.

First, the uncertainty about any underlying constitutional causes – as about much else concerning Brexit – makes it far from straightforward to anticipate how the future development of the UK constitution could respond to this agenda. Again, much would depend on the future substantive relationship agreed with the EU – if the UK were to end up formally outside of the Union, but within related structures such as the single market, and consequently subject to fields of EU law, it would be open to question whether this could amount to a restoration of sovereignty in any sense. Indeed, many might suggest that in such circumstances the UK’s power to control its own affairs would be diminished, given the loss of influence over EU law-making this would necessarily entail. Even if the UK were to negotiate a position entirely apart from the EU and its legal norms, there are other international institutions and power structures with which we would be required to engage. Will the accountability gap be any less apparent, for example, with respect to the World Trade Organisation, or the institutions which could be empowered under international trade agreements?Footnote ¹¹¹

It is difficult to avoid the conclusion that if Brexit is intended to herald a return to a romanticised constitutional past it will be a significant disappointment. A UK constitution founded on the principle of untrammelled national sovereignty is not just impossible under modern conditions, but the very notion is a myth – this much should be readily apparent in a tradition which is sceptical of the utility of formal limitations, while recognising that political constraint can provide the core of a constitutional framework. The unavailability of a constitution that is invulnerable to European influence does not mean abandoning all ideas of (parliamentary) sovereignty, appropriately understood, operating in a domestic context. It also does not mean that the broad constitutional concerns discussed here should be disregarded as the unalterable pillars on which a supranational legal order must rest. Rather, the UK (in common with others) will likely continue to grapple with the questions of what it means to be a nation state in a world of globalised authority and diffuse institutional arrangements, and how best to combine ideas of sovereignty and autonomy with the realities of power and influence. Perhaps engagement with these challenges might be advanced through reduced focus on how to reclaim ‘lost’ national power, and greater reflection on how states can use their positions to enhance the accountability and responsiveness of supranational systems of governance. Yet paradoxically, this constitutional soul searching may be a more difficult task for the UK outside the EU than it would have been on the inside.

Second, there are questions here of broader importance for the EU’s remaining member states too – should Brexit be treated as a function of UK constitutional exceptionalism, or does it give cause for more general reflection regarding the scope, ambition and framing of the European project? While integration of the power and responsibilities of national and European institutions continues, how can the fault line – which though perhaps blurred, certainly remains – between nation states and supranational structures best be navigated? Is the relocation of competence to the supranational level generating a sense of remoteness – both for those exercising power and those subject to it – which is pernicious, and serves to undermine the benefits which greater cooperation can offer? How can the accountability gap be bridged? Whether Brexit really does mark the ‘high watermark’ of Europeanisation, let alone of globalisation,Footnote ¹¹² very much remains to be seen. But it may (re-)expose significant challenges for those interested in the operation of some of the most fundamental of constitutional principles at the supranational level – and it is not clear whether the EU is really any better placed to respond to these questions as to its functioning and legitimacy than is the UK in relation to its own many and varied difficulties.

Both in the UK and in the EU, therefore, we can choose to treat Brexit as a momentary challenge, presenting a set of particular problems which can be addressed (no doubt imperfectly), and then moved beyond. Or we can view it as raising more fundamental constitutional questions about government and governance, democracy and accountability, power and sovereignty, the national and the supranational. Whether Brexit does or does not ultimately represent a critical challenge to constitutional pluralism,Footnote ¹¹³ or constitutional cosmopolitanism,Footnote ¹¹⁴ or even (if the limits of what withdrawal from the EU can achieve become apparent) constitutional parochialism, is clearly a matter for the future. Yet, for now, it may be most appropriate to treat Brexit as both a general and a specific constitutional challenge – there is little room for complacency in such a response.